SNP discovery in the transcriptome of white Pacific shrimp Litopenaeus vannamei by next generation sequencing

Abstract

The application of next generation sequencing technology has greatly facilitated high throughput single nucleotide polymorphism (SNP) discovery and genotyping in genetic research. In the present study, SNPs were discovered based on two transcriptomes of Litopenaeus vannamei (L. vannamei) generated from Illumina sequencing platform HiSeq 2000. One transcriptome of L. vannamei was obtained through sequencing on the RNA from larvae at mysis stage and its reference sequence was de novo assembled. The data from another transcriptome were downloaded from NCBI and the reads of the two transcriptomes were mapped separately to the assembled reference by BWA. SNP calling was performed using SAMtools. A total of 58,717 and 36,277 SNPs with high quality were predicted from the two transcriptomes, respectively. SNP calling was also performed using the reads of two transcriptomes together, and a total of 96,040 SNPs with high quality were predicted. Among these 96,040 SNPs, 5,242 and 29,129 were predicted as non-synonymous and synonymous SNPs respectively. Characterization analysis of the predicted SNPs in L. vannamei showed that the estimated SNP frequency was 0.21% (one SNP per 476 bp) and the estimated ratio for transition to transversion was 2.0. Fifty SNPs were randomly selected for validation by Sanger sequencing after PCR amplification and 76% of SNPs were confirmed, which indicated that the SNPs predicted in this study were reliable. These SNPs will be very useful for genetic study in L. vannamei, especially for the high density linkage map construction and genome-wide association studies.

Figure 1. Venn diagram of SNPs discovered using reads of the three datasets.

M represented SNPs discovered using reads from transcriptome of L.vannmei at developmental stage of mysis, P represented SNPs discovered using reads from transcriptome of L.vannmei at developmental stage of post larva. M+P represented SNPs discovered using the reads of the two transcriptomes together.

Figure 2. Statistics of read depth in each SNP in L. vannamei.

The horizontal axis represented the read depth of SNPs, the vertical axis represented the number of SNPs with the corresponding read depth. The average read depth was 44.

Figure 3. Statistics of minor allele frequency of total discovered SNPs in L. vannamei.

SNP number with MAF below 0.2 was too small to be observed in the column diagram.

Figure 4. The distribution of SNPs in unigenes.

The horizontal axis represented SNP numbers per unigene. The vertical axis represented the number of unigenes.

Figure 5. The frequencies of SNPs in unigenes.

The frequencies of SNPs in each unigene was calculated by dividing unigene length by SNPs number per unigene. Number of unigenes with SNP frequency over 0.030 was too small to be observed in the column diagram.

Figure 6. Gene ontology of all the annotated unigenes and unigenes with SNP frequency more than 0.014.

The blue column represented gene ontology of all unigenes containing SNPs. The red column represented gene ontology of unigenes with SNP frequency more than 0.014.

Figure 7. The top 20 KEGG pathway classification of assigned SNPs.

The horizontal axis represented KEGG pathway annotation. The vertical axis represented the number of SNPs assigned to the corresponding KEGG pathway.